In a hospital, some patient features are requiring higher cost to collect (e.g., body or brain scans) whereas others are not (e.g., temperature, blood pressure). Therefore, we have decided to our classification algorithm to predict whether a patient has a disease, and if the classifier is 75% confident that the patient has a disease, then we will do additional examinations to collect additional patient features. In this case, which classification methods do you recommend: logistics regression, decision tree, or naive Bayes? Justify your answer in one or two sentences.